Exodus study finds some change, little harm
Dr. Warren ThrockmortonDr. Warren Throckmorton's Weblog
- 2007 Sep 17
Last Thursday, September 13, in Nashville, Stanton Jones (Provost, Wheaton College) and Mark Yarhouse (Professor of Psychology, Regent University) presented the results of a study of religiously mediated change of sexual orientation. To a packed house, the researchers outlined the methods of sampling, the measures used and the results. Following the presentation, Intervarsity Press hosted a brief press conference.
Key points and findings:
The study sought to address two questions: Is change of sexual orientation, specifically homosexual orientation, possible? And, is the attempt intrinsically harmful? The authors were careful to point out that the participants were not engaged in professional therapy and so the variable of interest was participation in Exodus, not reparative therapy. Jones and Yarhouse began with 98 subjects and at time 3 assessment reported responses from the 73 remaining. The retention rate of 74.5% is respectable as compared with other longitudinal studies.
Using several measures of sexual orientation (including Kinsey scale, Klein scales, Shively and Dececco and self-report of categorical change), the authors reported change in several different ways. I’ll note three here. First of all, when simply asked how the participants thought of themselves, the results were as follows from Time 1 to Time 3 (over 4 years).
- 33 people reported change in the desired manner (from gay at time 1 in the heterosexual direction at time 3)
- 29 reported no change
- 8 reported change in the undesired direction
- 3 were unsure how to describe their experience of change
In addition, Jones and Yarhouse segmented a subgroup they called “Truly Gay.” This group was expected to show less change since they had more settled homosexual attractions, a gay identity and past homosexual activity. However, this group demonstrated a larger degree of change. Since multiple measures were used, it is difficult to summarize the degree of change they reported. However, I will report one example dimension here. For the entire population, a Kinsey self-rating was developed with one item used to inform the rating. For the whole population, an average rating of 5.07 was reported at Time 1 (the beginning). At time three, the average was 4.08, or almost a one point decline on average which is a significant result. Some people reported lots of change, others not so much as noted above. On average, the changes were statistically significant. However, observers might wonder if these changes are of a sufficient practical difference to warrant optimism about claims of change. My response is that even some change with little evidence of harm is of great importance to people who are seeking great congruence with their values and beliefs. The authors were quite careful to note that the changes reported were modest for most. They also noted that diminishment of homosexual attractions was more pronounced than acquisition of heterosexual attractions.
Other categories reported were:
- Success: Conversion - There were subjects who reported that they felt their change to be successful and reported substantial reduction in homosexual desire and addition of heterosexual attraction and functioning at Time 3. 15% met these criteria.
- Success: Chastity - These people experienced satisfactory reductions in homosexual desire and were living chaste lives. 23% were in this category.
- Continuing - These persons experienced only modest change in the desired direction but expressed commitment to continue. 29% were in this category.
- No-response - These people experienced no change and were conflicted about the future even though they had not given up. 15% were here.
- Failure (from their perspective): Confused - No change reported and had given up but did not label themselves gay. 4% were in this group
- Failure: Gay identity - No change, no pursuit and had come as gay. 8% were in this category.
Regarding harm, results of the Symptom Check List - 90 - Revised (SCL-90) were changed little from Time 1 to Time 3. The entire sample was in better mental health shape than existing norms for outpatients at Time 1 and improved slightly by Time 3.
The authors are to be commended for their candor and the tentative way of describing their results. They clearly noted the limitations and the strength and made appropriate qualifications. They were careful to acknowledge the reality of harm that can occur from poor practices and made no attempt to minimize the harm that has been reported (e.g., the ex-ex-gays). The book with lots of supporting material is available through Intervarsity Press.
Warren Throckmorton, PhD